A Risk Prediction Model for Progression-Free Survival in Endometrial Cancer Integrating Clinicopathological Variables...

Area of Science:

• Gynecologic Oncology
• Clinical Prediction Modeling
• Cancer Biomarkers

Background:

• Endometrial cancer (EC) prognosis is influenced by various factors.
• Accurate prediction of progression-free survival (PFS) is crucial for treatment planning.
• Existing models may not fully incorporate readily available clinical and laboratory data.

Purpose of the Study:

• To develop and validate a prediction model for PFS in EC patients.
• To assess the prognostic value of clinicopathological, inflammatory, nutritional, and tumor markers.
• To evaluate the incremental prognostic contribution of specific biomarkers.

Main Methods:

• Retrospective cohort study of 282 EC patients.
• Extraction of clinicopathological and preoperative laboratory data.
• Kaplan-Meier and Cox regression for prognostic factor identification and model construction.
• Internal validation using bootstrapping and performance metrics (C-index, ROC, calibration, DCA).

Main Results:

• FIGO stage III-IV, non-endometrioid histology, neutrophil-to-lymphocyte ratio (NLR), and cancer antigen 125 (CA125) were independent predictors of PFS.
• The developed model demonstrated good discrimination for 3-year PFS (AUC = 0.792).
• NLR and CA125 provided added prognostic value beyond traditional factors.

Conclusions:

• A prediction model using routine data shows good internal performance for EC PFS risk stratification.
• NLR and CA125 offer incremental prognostic insights, aiding personalized postoperative management.
• External validation is recommended prior to widespread clinical adoption.